PROJECT SUMMARY Our long-term goal is to develop a point-of-care urine test as a rapid, non-invasive surrogate for therapeutic drug monitoring of anti-tuberculosis drugs. The overall objective of this application is to generate receiver-operating- characteristic curves for urine colorimetry in the identification of tuberculosis patients with low anti-tuberculosis drug exposures. The study population will include children and adults with active tuberculosis. Our central hypothesis is that urine colorimetry will successfully identify patients with low anti-tuberculosis plasma drug levels, and do so with diagnostic characteristics similar to other widely used tests in tuberculosis management. The premise for this hypothesis is based on our work showing that urine colorimetry accurately identified adult tuberculosis patients with low anti- tuberculosis plasma drug exposures. This project includes two specific aims. In Aim 1, we will conduct a prospective, observational, intensive pharmacokinetic study of isoniazid, rifampin, and pyrazinamide among adults treated for active tuberculosis. We will sample both blood and urine in order to define the receiver-operating-characteristic curve for urine colorimetry to detect low anti-tuberculosis drug exposures, as defined by the gold standard measure of plasma drug exposure (the area under the 24-hour concentration-versus-time curve, estimated from the population pharmacokinetic model for each participant). Because of the demonstrated value of therapeutic drug monitoring in co-morbid tuberculosis/diabetes patients, we will stratify enrollment based on diabetes mellitus status. This stratification will support adjusted receiver- operating-characteristic analyses of the urine colorimetric assay with diabetes as a covariate. In Aim 2, we will determine the diagnostic characteristics of urine colorimetry to detect low plasma anti-tuberculosis drug exposures among children and adolescents with active tuberculosis disease in a high-burden setting (Tanzania), following a similar statistical analysis plan (pharmacokinetic analysis followed by receiver-operating- characteristic curves for urine colorimetry). The rationale for the proposed research is that once receiver- operating-characteristic curves for urine colorimetry are established among two key groups of high-risk tuberculosis patients (adults with co-morbid diabetes mellitus and children), translation of these assays to a point-of-care test can be performed. The urine colorimetric assay can then be validated as a noninvasive method for optimizing anti-tuberculosis drug exposures in resource-limited settings.